Background: Unanticipated outcomes and harms resulting from medical care are inevitable. Disclosure of those outcomes to patients is increasingly expected and required, but this practice remains uncommon and few physicians have had disclosure training. Failure to conduct disclosure consistently and well represents a fundamental threat to patient-centered care. Many have asserted that improving disclosure could increase patient satisfaction and reduce malpractice claims, but empirical evidence for these claims is lacking. No prior studies have assessed the quality of actual disclosures, hampering efforts to improve the disclosure process. Objectives: We propose a randomized trial of training physicians in disclosure with the following specific aims: 1) to determine whether physician communication training in disclosing unanticipated outcomes to patients affects patient satisfaction with disclosure;2) to explore whether physician communication training in disclosing unanticipated outcomes to patients affects malpractice claims;3) to explore whether characteristics of the event (severity of harm, presence of error), the physician, patient, and the environment independently affect the relationship between unanticipated outcome disclosure training and patient satisfaction. Methods: Participants will be 400 surgeons and interventional physicians in Colorado (who obtain liability insurance through COPIC) and in Washington (UW Medicine). The intervention includes: a) 2-hour disclosure training webcast;b) practice and feedback with two standardized patients;and c) a refresher training webcast mid-project. The primary outcome will be patient satisfaction with a disclosure they experience, as measured with the Patient Assessment of Disclosure Quality, a psychometrically sound instrument. The secondary outcome will be the incidence of malpractice claims in the subsequent 24 months. Multivariate regression analysis will also be used to examine the effects of patient, event, and physician characteristics on patient satisfaction with disclosure. Implications and Impact: This study will be the first to measure the quality of actual disclosures and assess whether physician training improves the disclosure process. The proposed project brings together leading researchers in medical error, malpractice, and communication skills training with two industry partners--a major malpractice insurance company and TMIT, a leader in web-based safety education. The findings will be widely applicable to training physicians to meet the increasing demands for effective disclosure.